The main objective of the project is the development of an assistance system using artificial vision technology applied to the non-invasive clinical protocol for embryo selection.

Fertility impairment affects approximately 80 million people worldwide. Although infertility rates vary widely, ranging from less than 5% to more than 30%, it is estimated that one in ten couples suffers from primary or secondary infertility problems. To cope with this problem, Assisted Reproduction Techniques (ART) were developed. The introduction of time-lapse incubators allowed continuous monitoring of the embryo, responding to the limitations of conventional evaluation by facilitating the study of embryo development without disturbing the culture conditions.

Currently, embryos are manually assessed by an expert embryologist who studies the morphology at different stages of development and classifies them into predetermined classes. However, this method of manual annotation is time-consuming and highly susceptible to inter- and intra-observer variability, making it subjective and inaccurate.

This project arises from the need to implement personalized clinical protocols that help increase the success of each assisted reproduction cycle and reduce the number of attempts to achieve reproductive success. Therefore, the aim is to establish a new more robust computer vision technology as part of a non-invasive clinical protocol capable of automatically selecting the embryo with the highest probability of implantation from a cohort.